V10 of modified SETI MB CUDA + opt AP package for full multi-GPU+CPU use

Message boards : Number crunching : V10 of modified SETI MB CUDA + opt AP package for full multi-GPU+CPU use
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 8 · 9 · 10 · 11 · 12 · 13 · 14 . . . 15 · Next

AuthorMessage
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 885361 - Posted: 14 Apr 2009, 21:29:33 UTC - in response to Message 885360.  

question to you guys.

does it impact overall performance of cuda app if i lower
this two values <avg_ncpus>0.127970</avg_ncpus>; <max_ncpus>0.127970</max_ncpus> to something like 0.03-0.04 ?

any answer/opinion is apreciated

thx

It's just hint to BOINC scheduler, not any realy CPU usage or CPU usage limit.
ID: 885361 · Report as offensive
piper69

Send message
Joined: 25 Sep 08
Posts: 49
Credit: 3,042,244
RAC: 0
Romania
Message 885362 - Posted: 14 Apr 2009, 21:34:31 UTC
Last modified: 14 Apr 2009, 21:37:20 UTC

problem is when i look in task manager the cuda app is constantly doing around 12% cpu usage.


is that normal?
ID: 885362 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 885485 - Posted: 15 Apr 2009, 8:04:40 UTC - in response to Message 885362.  

problem is when i look in task manager the cuda app is constantly doing around 12% cpu usage.


is that normal?

It depends on CPU speed. I looked at your Athlon 64 X2 host results - it seems CUDA part working OK. You could see big CPU usage for CUDA app when it can't properly initialize (low GPU memory or another error) and falls back to CPU processing. This mode should be avoided cause CPU processing by CUDA MB much slower than by AK_v8. But your GPU works OK.
ID: 885485 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 885506 - Posted: 15 Apr 2009, 9:01:12 UTC - in response to Message 884987.  
Last modified: 15 Apr 2009, 9:01:52 UTC

...
Since BOINC V6.6.20 your nice team mod isn't needed anymore.
Maybe you can open a new thread with the latest available CUDA versions? [V10/11]
[incl. dll's and needed app_info.xml]

Or maybe a hint to the app_info.xml thread?
app_info for AP500, AP503, MB603 and MB608


Which/where are the .dll-versions from you?
Please could you post where to find?
Or this are the .dll's from your V7 and V10 mod. This are the only different .dll-versions from you?
[in the 1st posts of the threads @ lunatics.kwsn.net?]

I could use the stock Berkeley .dll's also with your mod?
Or what's different?
I have no knowledge for what the .dll's are needed.. I'm not the profi.. I'm an 'amateur'.. ;-)


You didn't saw my questions? ;-)


If you don't have time with a new thread..
Could you please post the URLs of the recommend CUDA Vx apps and .dll's?

Then someone of the community could open a new thread because of this.. :-)

ID: 885506 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 885517 - Posted: 15 Apr 2009, 10:02:00 UTC - in response to Message 885506.  

ID: 885517 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 885627 - Posted: 15 Apr 2009, 19:16:32 UTC - in response to Message 885517.  

Could you please post the URLs of the recommend CUDA Vx apps and .dll's?

http://lunatics.kwsn.net/12-gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg16715.html#msg16715


Thanks a lot!


Hmm.. but.. what about the rigs with > 1 GPU in it..?

Soon also a new app release for this rigs?

ID: 885627 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 885831 - Posted: 16 Apr 2009, 16:44:59 UTC - in response to Message 885627.  

Could you please post the URLs of the recommend CUDA Vx apps and .dll's?

http://lunatics.kwsn.net/12-gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg16715.html#msg16715


Thanks a lot!


Hmm.. but.. what about the rigs with > 1 GPU in it..?

Soon also a new app release for this rigs?


..hmm.. I misread your post at lunatics.kwsn.net ?
I understand it, that this app is ONLY for rigs with 1 GPU.

No?

ID: 885831 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 885851 - Posted: 16 Apr 2009, 18:14:52 UTC - in response to Message 885831.  

Could you please post the URLs of the recommend CUDA Vx apps and .dll's?

http://lunatics.kwsn.net/12-gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg16715.html#msg16715


Thanks a lot!


Hmm.. but.. what about the rigs with > 1 GPU in it..?

Soon also a new app release for this rigs?


..hmm.. I misread your post at lunatics.kwsn.net ?
I understand it, that this app is ONLY for rigs with 1 GPU.

No?

You read it correct.
I updated that post and included builds w/o affinity lock based on current codebase. No actual performance difference should be with last V10/11 update buried in corresponding Lunatic's thread. I did fresh build just for convenience.

ID: 885851 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 886053 - Posted: 17 Apr 2009, 7:31:52 UTC - in response to Message 885851.  

Build w/o VLAR kill mod added to the same post. It should load faster (use less CPU time) than current stock while doing all kind of work stock does.
http://lunatics.kwsn.net/12-gpu-crunching/v10-of-modified-seti-mb-cuda-opt-ap-package-for-full-multi-gpucpu-use.msg16715.html#msg16715
(remember that VLAR and VHAR tasks use GPU less effective than midrange tasks, complete data chart in progress).
ID: 886053 · Report as offensive
Profile Voyager
Volunteer tester
Avatar

Send message
Joined: 2 Nov 99
Posts: 602
Credit: 3,264,813
RAC: 0
United States
Message 886095 - Posted: 17 Apr 2009, 21:14:19 UTC
Last modified: 17 Apr 2009, 21:23:15 UTC

Just noticed my gpu temp dropped quite a bit,shows it's running in task manager,but in boinc mgr times went from ~14min to 30min now one running at about 2hrs. I'll do a reboot now and see what happens.
edit..After reboot temp went back to normal, also processing time looks regular./edit
ID: 886095 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 886101 - Posted: 17 Apr 2009, 21:46:51 UTC - in response to Message 886095.  

ID: 886101 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 887306 - Posted: 22 Apr 2009, 15:51:46 UTC - in response to Message 886101.  

bump keeping this up so cuda users can find it


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 887306 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 887319 - Posted: 22 Apr 2009, 16:34:48 UTC - in response to Message 885265.  


http://setiathome.berkeley.edu/result.php?resultid=1195495666
icfft=86040, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error

http://setiathome.berkeley.edu/result.php?resultid=1190297258
icfft=94365, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error

http://setiathome.berkeley.edu/result.php?resultid=1195160891
icfft=86665, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error

http://setiathome.berkeley.edu/result.php?resultid=1202303914
icfft=84509, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error

http://setiathome.berkeley.edu/result.php?resultid=1202338717
icfft=98384, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error


Two more:

http://setiathome.berkeley.edu/result.php?resultid=1205398993
icfft=84915, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error

http://setiathome.berkeley.edu/result.php?resultid=1205430734
icfft=86092, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error


---------------------------------------------------------------

If you would like to help Raistmer.. please post the '-12 Unknown error' also.. *thumb up*

It could look like this:
Exception detected inside cudaAcc_find_triplets, dumping client state
icfft=98384, PoT_activity=0, PoT_freq_bin=-1SETI@home error -12 Unknown error
cudaAcc_find_triplets erroneously found a triplet twice in find_triplets_kernel
File: ..\analyzePoT.cpp
Line: 348


And only the [bolded] line is needed.


ID: 887319 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 887520 - Posted: 23 Apr 2009, 8:48:14 UTC


I'm curious..

I had a bunch of VLAR 'killed' WUs.. [52]
And BOINC uploaded them like every time..

Then BOINC had a ~ 24 hours break counter in the project overview. [BOINC V6.4.7]
Why?


BTW.
Why the 'killed' VLARs need to upload?

ID: 887520 · Report as offensive
Profile Virtual Boss*
Volunteer tester
Avatar

Send message
Joined: 4 May 08
Posts: 417
Credit: 6,440,287
RAC: 0
Australia
Message 887524 - Posted: 23 Apr 2009, 9:08:51 UTC - in response to Message 887520.  
Last modified: 23 Apr 2009, 9:11:32 UTC

The answer is in This thread

BTW answer - To let the server know that you are aborting them, otherwise they would have to time out (no response).

Editted for correct thread
ID: 887524 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 888339 - Posted: 25 Apr 2009, 21:17:24 UTC
Last modified: 25 Apr 2009, 21:18:56 UTC


http://setiathome.berkeley.edu/forum_thread.php?id=52212&nowrap=true#884686

Because of my 'Out Of Memory' errors..

Some examples:
http://setiathome.berkeley.edu/result.php?resultid=1206226503
http://setiathome.berkeley.edu/result.php?resultid=1211721515
http://setiathome.berkeley.edu/result.php?resultid=1212504106


All errors look like this:
-------------------------------------------------------
WU true angle range is : x.xxxxxx


Unhandled Exception Detected...

- Unhandled Exception Record -
Reason: Out Of Memory (C++ Exception) (0xe06d7363)* at address 0x7C81EB33

Engaging BOINC Windows Runtime Debugger...



********************


BOINC Windows Runtime Debugger Version 6.3.22
-------------------------------------------------------



I guess now it's not a RAM error it's looks like it's a compiler error.
A german side:
* http://support.microsoft.com/kb/185294/de

Or not?

ID: 888339 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 888353 - Posted: 25 Apr 2009, 22:30:45 UTC - in response to Message 888339.  

Compiler error? Don't think so.
Maybe it's worth to add dump of available system physical memory too in additional to GPU memory. Will think about it.
For now such out of memory behavior could be summoned by buggy stock 6.03 on beta (if I correctly understood Pappa's report).
ID: 888353 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 888363 - Posted: 25 Apr 2009, 22:56:49 UTC


This '0x7C81EB33' couldn't be the address of the place of a damaged RAM, or?

Like I said, the error output is all the time the same.

ID: 888363 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 888369 - Posted: 25 Apr 2009, 23:26:25 UTC - in response to Message 888339.  

"at address 0x7C81EB33"
This adress the same? Same exception code means same error - out of memory.

ID: 888369 · Report as offensive
Profile Dirk Sadowski
Volunteer tester

Send message
Joined: 6 Apr 07
Posts: 7105
Credit: 147,663,825
RAC: 5
Germany
Message 888417 - Posted: 26 Apr 2009, 6:43:10 UTC


Yes, all the time the same output:

------------------------------------------------
Unhandled Exception Detected...

- Unhandled Exception Record -
Reason: Out Of Memory (C++ Exception) (0xe06d7363) at address 0x7C81EB33

Engaging BOINC Windows Runtime Debugger...

------------------------------------------------


I'm confused - I'm the only one which see this error in the own overview? ;-)


ID: 888417 · Report as offensive
Previous · 1 . . . 8 · 9 · 10 · 11 · 12 · 13 · 14 . . . 15 · Next

Message boards : Number crunching : V10 of modified SETI MB CUDA + opt AP package for full multi-GPU+CPU use


 
©2025 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.